This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Project Summary Grant Number: R21-EB005690 Project Start: 14-JUL-2006 Project Initial End: 30-JUN-2008 Project End: 30-JUN-2009 The goal of this project is to improve the quality and spatial accuracy of echo-planar imaging (EPI), so that accurate quantitative information can be derived from EPI based medical research and clinical diagnosis. EPI is one of the fastest MR imaging techniques, and has been popularly applied to various dynamic studies that require high temporal-resolution, such as functional MRI (fMRI), contrast-enhanced imaging, and MR based interventional procedures. However, EPI data quality is usually degraded by various artifacts, such as geometric distortions and susceptibility signal loss. Furthermore, the sensitivity of EPI to susceptibility field nhomogeneities makes it less reliable in EPI based longitudinal studies. Several techniques have been previously reported for EPI quality improvement and artifact reduction. However, most previously reported EPI artifact reduction methods require time-consuming field mapping scans, and therefore may not always be practical (e.g. for clinical scans and EPI based interventional MRI procedures). Here we propose to use a novel k-space energy spectrum analysis to quantify (1) the k-space energy distribution, (2) susceptibility field gradients, (3) the spatially-dependent echo time values, and (4) artifact levels directly from the acquired EPI data, without the need of additional field mapping procedure or pulse sequence modification. Various EPI artifacts (e.g. distortions and Gibb's ripple artifact) can be effectively removed using the proposed approach. Furthermore, the developed k-space energy spectrum analysis will be applied to design an optimal acquisition strategy for phase-encoded 3D parallel EPI, with an improved signal-to-noise ratio and reduced motion related artifact. We also plan to apply the proposed methods to re-analyze the previously acquired fMRI data, and retrospectively improve the longitudinal reproducibility of grouped activation. The methods developed in the proposed project will be made available to MRI community so that other research groups may use the developed methods to improve their future EPI based quantitative studies or to retrospectively improve the EPI data that were previously obtained. Benefit to NCIGT The KESA method can be applied to provide robust temperature maps using EPII through reliable phase unwrapping procedure to effectively eliminate the phase wraparounds in dynamic temperature mapping. Based on our k-space energy spectrum analysis algorithm (R21 project), a new phase mapping and unwrapping method is being designed. The MRI based temperature mapping will have a better tolerance to subject movement and susceptibility effect when the new reliable phase unwrapping procedure is included. Thus, this project supports our work in the development of new temperature mapping methods. Field maps using the KESA method can be used for distortion correction in EPI-base fMRI and DTI. Thus this work supports our efforts in the Neurosurgery Core. Benefit to the Project The IGT resource provides an essential support to the R21 project. The programming environment for development of the sequence is maintained (in part) through support from the imaging core of the resource. In addition, a training fellow attached to the IGT resource, Ming-Long Wu, is participating in the experimental work of the R21 project.